DESCRIPTION: The overall aim of the project is to determine the structure of the adenovirus virion in atomic detail. A novel approach is mandated by its large size (ca. 1000 A diameter) and the complicated arrangement of its constituent 2,700 polypeptides from 10 different protein species. The human type 2 virion will be investigated by combining information from an X-ray crystallographic analysis of the coat protein hexon, with information from complementary electron microscopy studies. Hexon is a trimeric molecule with three identical polypeptide chains of 967 residues (109,077 Daltons each). The structure of adenovirus type 2 hexon is being completed at 2.2 A resolution with crystallographic methods. The hexon structure will be analyzed to uncover the contribution of charge, hydrogen bonding, hydropathy, and topology to the extraordinary stability of the molecule. The effect of pH on the structure will be studied to understand the basis for conformational changes that may be important in viral infectivity. The approximate positions of the hexon in the virion have been determined from difference imaging combining the crystallographic hexon model with an image reconstruction of the virion at 35 A resolution from cryo- electron microscopy. These positions will be refined using the 2.2 A hexon model to provide a high resolution atomic model for the 240 hexons in the capsid. This will permit a complete analysis of the hexon-hexon capsid interactions. The nature of the binding "pockets" of the minor capsid proteins will be investigated by comparing the high-resolution capsid model with the density for these proteins in the image reconstruction. This will provide a detailed view of the hexon environment for polypeptides IIIa, VI and IX, which bind at strategic positions and stabilize the virion. The chemical characteristics of the pockets will be used to model interactions of the minor proteins with hexon. A full understanding of the role played by "cementing" proteins not only has relevance for the architecture of adenovirus but also for their role in other large macromolecular assemblies. This will lay the foundation for drugs designed to disrupt the infective process by destabilizing or over-stabilizing the virion. The structure of the immunologically distinct human type 5 hexon will be determined, and the study extended to other serotypes. Comparative studies will reveal the structural basis for their immunological differences, and reveal how the viral coat has differentiated. They should also give a better understanding of the tissue-specificity of adenovirus serotypes, which has immediate relevance to the use of adenovirus as a vector in human gene therapy. The structural studies on adenovirus serve as a model for mammalian viruses and other assemblies that are too large to be amenable to crystallography alone. Study of these biological systems requires an integration of investigative methods including electron microscopy, biochemistry, and molecular biology to probe their full range of physical and biological properties.